In the past year, this project has addressed several problems and methodological challenges in protein folding, most significantly: 1. Methodology for protein folding simulations. A description of protein folding or misfolding using atomistic simulations relies on the quality of the simulation model. Despite recent successes in ab initio folding simulations, there are still outstanding problems with the simulation models. These have been addressed here at a number of levels. At the most detailed, we have contributed to the development of the next generation of polarizable energy functions, which in principle should provide the most accurate description of the protein energy landscape (12). We have also worked to develop simpler, but more inexpensive computationally methods: firstly, by parametrizing non-polarizable models to match polarizable ones (2) and secondly, by developing models in which the solvent does not need to be explicitly modelled (7). These advances should ultimately result in a more reliable description of protein dynamics and function. 2. Protein folding mechanism. We have used molecular simulation models to test current protein folding theory. This has included a direct test of some key assumptions of the theory against a set of detailed atomistic simulations of nine different proteins in explicit water. We have found that the assumptions of the theory hold for naturally occurring proteins, the only exceptions being for simplified, designed proteins (6). We have further tested the theory against experimental data, by using the assumptions of the theory in simulations to predict the folding mechanism of spectrin domains (5). These results help to justify the use of simplified molecular models to provide insight into protein folding and function (10). 3. Role of molecular chaperonins in protein folding. We have used simulation and theory in order to analyze the effect of passive confinement inside chaperonins such as GroEL on protein folding. We have been able to explain experimental observations on folding rates inside chaperonins, and to propose a specific mechanism by which chaperonins can prevent misfolding in multidomain proteins.